Face Generation¶

In this project, you'll use generative adversarial networks to generate new images of faces.

Get the Data¶

You'll be using two datasets in this project:

MNIST
CelebA

Since the celebA dataset is complex and you're doing GANs in a project for the first time, we want you to test your neural network on MNIST before CelebA. Running the GANs on MNIST will allow you to see how well your model trains sooner.

If you're using FloydHub, set data_dir to "/input" and use the FloydHub data ID "R5KrjnANiKVhLWAkpXhNBe".

data_dir = './data'

# FloydHub - Use with data ID "R5KrjnANiKVhLWAkpXhNBe"
#data_dir = '/input'


"""
DON'T MODIFY ANYTHING IN THIS CELL
"""
import helper

helper.download_extract('mnist', data_dir)
helper.download_extract('celeba', data_dir)

Found mnist Data
Found celeba Data

Explore the Data¶

MNIST¶

As you're aware, the MNIST dataset contains images of handwritten digits. You can view the first number of examples by changing show_n_images.

show_n_images = 25

"""
DON'T MODIFY ANYTHING IN THIS CELL
"""
%matplotlib inline
import os
from glob import glob
from matplotlib import pyplot

mnist_images = helper.get_batch(glob(os.path.join(data_dir, 'mnist/*.jpg'))[:show_n_images], 28, 28, 'L')
pyplot.imshow(helper.images_square_grid(mnist_images, 'L'), cmap='gray')

<matplotlib.image.AxesImage at 0x184c230beb8>

---------------------------------------------------------------------------
AttributeError                            Traceback (most recent call last)
C:\anaconda\envs\tensorflow35\lib\site-packages\IPython\core\formatters.py in __call__(self, obj)
    330                 pass
    331             else:
--> 332                 return printer(obj)
    333             # Finally look for special method names
    334             method = get_real_method(obj, self.print_method)

C:\anaconda\envs\tensorflow35\lib\site-packages\IPython\core\pylabtools.py in <lambda>(fig)
    235 
    236     if 'png' in formats:
--> 237         png_formatter.for_type(Figure, lambda fig: print_figure(fig, 'png', **kwargs))
    238     if 'retina' in formats or 'png2x' in formats:
    239         png_formatter.for_type(Figure, lambda fig: retina_figure(fig, **kwargs))

C:\anaconda\envs\tensorflow35\lib\site-packages\IPython\core\pylabtools.py in print_figure(fig, fmt, bbox_inches, **kwargs)
    119 
    120     bytes_io = BytesIO()
--> 121     fig.canvas.print_figure(bytes_io, **kw)
    122     data = bytes_io.getvalue()
    123     if fmt == 'svg':

C:\anaconda\envs\tensorflow35\lib\site-packages\matplotlib\backend_bases.py in print_figure(self, filename, dpi, facecolor, edgecolor, orientation, format, **kwargs)
   2206                     orientation=orientation,
   2207                     dryrun=True,
-> 2208                     **kwargs)
   2209                 renderer = self.figure._cachedRenderer
   2210                 bbox_inches = self.figure.get_tightbbox(renderer)

C:\anaconda\envs\tensorflow35\lib\site-packages\matplotlib\backends\backend_agg.py in print_png(self, filename_or_obj, *args, **kwargs)
    505 
    506     def print_png(self, filename_or_obj, *args, **kwargs):
--> 507         FigureCanvasAgg.draw(self)
    508         renderer = self.get_renderer()
    509         original_dpi = renderer.dpi

C:\anaconda\envs\tensorflow35\lib\site-packages\matplotlib\backends\backend_agg.py in draw(self)
    428             if toolbar:
    429                 toolbar.set_cursor(cursors.WAIT)
--> 430             self.figure.draw(self.renderer)
    431         finally:
    432             if toolbar:

C:\anaconda\envs\tensorflow35\lib\site-packages\matplotlib\artist.py in draw_wrapper(artist, renderer, *args, **kwargs)
     53                 renderer.start_filter()
     54 
---> 55             return draw(artist, renderer, *args, **kwargs)
     56         finally:
     57             if artist.get_agg_filter() is not None:

C:\anaconda\envs\tensorflow35\lib\site-packages\matplotlib\figure.py in draw(self, renderer)
   1293 
   1294             mimage._draw_list_compositing_images(
-> 1295                 renderer, self, artists, self.suppressComposite)
   1296 
   1297             renderer.close_group('figure')

C:\anaconda\envs\tensorflow35\lib\site-packages\matplotlib\image.py in _draw_list_compositing_images(renderer, parent, artists, suppress_composite)
    136     if not_composite or not has_images:
    137         for a in artists:
--> 138             a.draw(renderer)
    139     else:
    140         # Composite any adjacent images together

C:\anaconda\envs\tensorflow35\lib\site-packages\matplotlib\artist.py in draw_wrapper(artist, renderer, *args, **kwargs)
     53                 renderer.start_filter()
     54 
---> 55             return draw(artist, renderer, *args, **kwargs)
     56         finally:
     57             if artist.get_agg_filter() is not None:

C:\anaconda\envs\tensorflow35\lib\site-packages\matplotlib\axes\_base.py in draw(self, renderer, inframe)
   2397             renderer.stop_rasterizing()
   2398 
-> 2399         mimage._draw_list_compositing_images(renderer, self, artists)
   2400 
   2401         renderer.close_group('axes')

C:\anaconda\envs\tensorflow35\lib\site-packages\matplotlib\image.py in _draw_list_compositing_images(renderer, parent, artists, suppress_composite)
    136     if not_composite or not has_images:
    137         for a in artists:
--> 138             a.draw(renderer)
    139     else:
    140         # Composite any adjacent images together

C:\anaconda\envs\tensorflow35\lib\site-packages\matplotlib\artist.py in draw_wrapper(artist, renderer, *args, **kwargs)
     53                 renderer.start_filter()
     54 
---> 55             return draw(artist, renderer, *args, **kwargs)
     56         finally:
     57             if artist.get_agg_filter() is not None:

C:\anaconda\envs\tensorflow35\lib\site-packages\matplotlib\image.py in draw(self, renderer, *args, **kwargs)
    546         else:
    547             im, l, b, trans = self.make_image(
--> 548                 renderer, renderer.get_image_magnification())
    549             if im is not None:
    550                 renderer.draw_image(gc, l, b, im)

C:\anaconda\envs\tensorflow35\lib\site-packages\matplotlib\image.py in make_image(self, renderer, magnification, unsampled)
    772         return self._make_image(
    773             self._A, bbox, transformed_bbox, self.axes.bbox, magnification,
--> 774             unsampled=unsampled)
    775 
    776     def _check_unsampled_image(self, renderer):

C:\anaconda\envs\tensorflow35\lib\site-packages\matplotlib\image.py in _make_image(self, A, in_bbox, out_bbox, clip_bbox, magnification, unsampled, round_to_pixel_border)
    410 
    411                 mask = np.empty(A.shape, dtype=np.float32)
--> 412                 if A.mask.shape == A.shape:
    413                     # this is the case of a nontrivial mask
    414                     mask[:] = np.where(A.mask, np.float32(np.nan),

AttributeError: 'numpy.ndarray' object has no attribute 'mask'

<matplotlib.figure.Figure at 0x184d02ac390>

CelebA¶

The CelebFaces Attributes Dataset (CelebA) dataset contains over 200,000 celebrity images with annotations. Since you're going to be generating faces, you won't need the annotations. You can view the first number of examples by changing show_n_images.

show_n_images = 25

"""
DON'T MODIFY ANYTHING IN THIS CELL
"""
mnist_images = helper.get_batch(glob(os.path.join(data_dir, 'img_align_celeba/*.jpg'))[:show_n_images], 28, 28, 'RGB')
pyplot.imshow(helper.images_square_grid(mnist_images, 'RGB'))

<matplotlib.image.AxesImage at 0x184a6ede550>

Preprocess the Data¶

Since the project's main focus is on building the GANs, we'll preprocess the data for you. The values of the MNIST and CelebA dataset will be in the range of -0.5 to 0.5 of 28x28 dimensional images. The CelebA images will be cropped to remove parts of the image that don't include a face, then resized down to 28x28.

The MNIST images are black and white images with a single color channel while the CelebA images have 3 color channels (RGB color channel).

Build the Neural Network¶

You'll build the components necessary to build a GANs by implementing the following functions below:

model_inputs
discriminator
generator
model_loss
model_opt
train

Check the Version of TensorFlow and Access to GPU¶

This will check to make sure you have the correct version of TensorFlow and access to a GPU

"""
DON'T MODIFY ANYTHING IN THIS CELL
"""
from distutils.version import LooseVersion
import warnings
import tensorflow as tf

# Check TensorFlow Version
assert LooseVersion(tf.__version__) >= LooseVersion('1.0'), 'Please use TensorFlow version 1.0 or newer.  You are using {}'.format(tf.__version__)
print('TensorFlow Version: {}'.format(tf.__version__))

# Check for a GPU
if not tf.test.gpu_device_name():
    warnings.warn('No GPU found. Please use a GPU to train your neural network.')
else:
    print('Default GPU Device: {}'.format(tf.test.gpu_device_name()))

TensorFlow Version: 1.1.0

C:\anaconda\envs\tensorflow35\lib\site-packages\ipykernel_launcher.py:14: UserWarning: No GPU found. Please use a GPU to train your neural network.

Input¶

Implement the model_inputs function to create TF Placeholders for the Neural Network. It should create the following placeholders:

Real input images placeholder with rank 4 using image_width, image_height, and image_channels.
Z input placeholder with rank 2 using z_dim.
Learning rate placeholder with rank 0.

Return the placeholders in the following the tuple (tensor of real input images, tensor of z data)

import problem_unittests as tests

def model_inputs(image_width, image_height, image_channels, z_dim):
    """
    Create the model inputs
    :param image_width: The input image width
    :param image_height: The input image height
    :param image_channels: The number of image channels
    :param z_dim: The dimension of Z
    :return: Tuple of (tensor of real input images, tensor of z data, learning rate)
    """
    # TODO: Implement Function
    real_imgs = tf.placeholder(tf.float32, (None, image_width, image_height, image_channels), 'real_imgs')
    z_data = tf.placeholder(tf.float32, (None, z_dim), 'z_data')
    learning_rate = tf.placeholder(tf.float32, name='learning_rate')
    return real_imgs, z_data, learning_rate


"""
DON'T MODIFY ANYTHING IN THIS CELL THAT IS BELOW THIS LINE
"""
tests.test_model_inputs(model_inputs)

Tests Passed

Discriminator¶

Implement discriminator to create a discriminator neural network that discriminates on images. This function should be able to reuse the variables in the neural network. Use tf.variable_scope with a scope name of "discriminator" to allow the variables to be reused. The function should return a tuple of (tensor output of the discriminator, tensor logits of the discriminator).

stride = 5
padding = 2
img_filter_size = 64
h2_img_filter_size = img_filter_size*2
h3_img_filter_size = img_filter_size*4


def discriminator(images, reuse=False):
    """
    Create the discriminator network
    :param images: Tensor of input image(s)
    :param reuse: Boolean if the weights should be reused
    :return: Tuple of (tensor output of the discriminator, tensor logits of the discriminator)
    """
    with tf.variable_scope('discriminator', reuse=reuse):
        # No alpha given so making a test        
        alpha = 0.2      
        
        # 1st Hidden layer
        h1 = tf.layers.conv2d(images, img_filter_size, stride, padding, 'same')
        h1 = tf.maximum(alpha * h1, h1) # Leaky ReLU
        
        # 2nd Hidden layer
        h2 = tf.layers.conv2d(h1, h2_img_filter_size, stride, padding, 'same')
        h2 = tf.layers.batch_normalization(h2, training=True) # Batch normalizing
        h2 = tf.maximum(alpha * h2, h2) # Leaky ReLU
        
        # 3rd Hidden layer
        h3 = tf.layers.conv2d(h2, h3_img_filter_size, stride, padding, 'same')
        h3 = tf.layers.batch_normalization(h3, training=True) # Batch normalizing
        h3 = tf.maximum(alpha * h3, h3) # Leaky ReLU
        
        flat = tf.reshape(h3, (-1, 4*4*h3_img_filter_size))
        logits = tf.layers.dense(flat, 1)
        out = tf.sigmoid(logits)
        
    return out, logits


"""
DON'T MODIFY ANYTHING IN THIS CELL THAT IS BELOW THIS LINE
"""
tests.test_discriminator(discriminator, tf)

Tests Passed

Generator¶

Implement generator to generate an image using z. This function should be able to reuse the variables in the neural network. Use tf.variable_scope with a scope name of "generator" to allow the variables to be reused. The function should return the generated 28 x 28 x out_channel_dim images.

def generator(z, out_channel_dim, is_train=True):
    """
    Create the generator network
    :param z: Input z
    :param out_channel_dim: The number of channels in the output image
    :param is_train: Boolean if generator is being used for training
    :return: The tensor output of the generator
    """
    # TODO: Implement Function
    with tf.variable_scope('generator', reuse=not is_train):
        # Using the reverse process of the discriminator
        alpha = 0.2
    
        h1 = tf.layers.dense(z, 2*2*h3_img_filter_size)
        h1 = tf.reshape(h1, (-1, 2, 2, h3_img_filter_size))
        h1 = tf.layers.batch_normalization(h1, training=is_train)
        h1 = tf.maximum(alpha * h1, h1)
    
        h2 = tf.layers.conv2d_transpose(h1, h2_img_filter_size, 5, 2, 'valid')
        h2 = tf.layers.batch_normalization(h2, training=is_train)
        h2 = tf.maximum(alpha * h2, h2)
    
        h3 = tf.layers.conv2d_transpose(h2, img_filter_size, 5, 2, 'same')
        h3 = tf.layers.batch_normalization(h3, training=is_train)
        h3 = tf.maximum(alpha * h3, h3)
    
        logits = tf.layers.conv2d_transpose(h3, out_channel_dim, 5, 2, 'same')
        out = tf.tanh(logits)
    
        return out


"""
DON'T MODIFY ANYTHING IN THIS CELL THAT IS BELOW THIS LINE
"""
tests.test_generator(generator, tf)

Tests Passed

Loss¶

Implement model_loss to build the GANs for training and calculate the loss. The function should return a tuple of (discriminator loss, generator loss). Use the following functions you implemented:

discriminator(images, reuse=False)
generator(z, out_channel_dim, is_train=True)

def model_loss(input_real, input_z, out_channel_dim):
    """
    Get the loss for the discriminator and generator
    :param input_real: Images from the real dataset
    :param input_z: Z input
    :param out_channel_dim: The number of channels in the output image
    :return: A tuple of (discriminator loss, generator loss)
    """
    g_model = generator(input_z, out_channel_dim)
    d_model_real, d_logits_real = discriminator(input_real)
    d_model_fake, d_logits_fake = discriminator(g_model, reuse=True)
    
    # Calculate the Losses for it real, fake & generated models
    d_loss_real = tf.reduce_mean(
        tf.nn.sigmoid_cross_entropy_with_logits(logits=d_logits_real, labels=tf.ones_like(d_model_real)))
    d_loss_fake = tf.reduce_mean(tf.nn.sigmoid_cross_entropy_with_logits(
        logits=d_logits_fake, labels=tf.zeros_like(d_model_fake)))    
    g_loss = tf.reduce_mean(tf.nn.sigmoid_cross_entropy_with_logits(
        logits=d_logits_fake, labels=tf.ones_like(d_model_fake)))
    
    # The Discriminator Loss is the amount of it fake img loss + it real img loss
    d_loss = d_loss_real + d_loss_fake
    return d_loss, g_loss


"""
DON'T MODIFY ANYTHING IN THIS CELL THAT IS BELOW THIS LINE
"""
tests.test_model_loss(model_loss)

Tests Passed

Optimization¶

Implement model_opt to create the optimization operations for the GANs. Use tf.trainable_variables to get all the trainable variables. Filter the variables with names that are in the discriminator and generator scope names. The function should return a tuple of (discriminator training operation, generator training operation).

def model_opt(d_loss, g_loss, learning_rate, beta1):
    """
    Get optimization operations
    :param d_loss: Discriminator loss Tensor
    :param g_loss: Generator loss Tensor
    :param learning_rate: Learning Rate Placeholder
    :param beta1: The exponential decay rate for the 1st moment in the optimizer
    :return: A tuple of (discriminator training operation, generator training operation)
    """
    # TODO: Implement Function
    # Empty TF TrainableVariables
    t_vars = tf.trainable_variables()
    # Append var to D & G vars arraies if it starts with the "prefix" for on it name
    d_vars = [var for var in t_vars if var.name.startswith('discriminator')]
    g_vars = [var for var in t_vars if var.name.startswith('generator')]
    
    # discriminator optimization
    d_train_opt = tf.train.AdamOptimizer(learning_rate, beta1=beta1).minimize(d_loss, var_list=d_vars)
    g_train_opt = tf.train.AdamOptimizer(learning_rate).minimize(g_loss, var_list=g_vars)
    
    # UPDATE OPS GraphKeys
    ops = tf.get_collection(tf.GraphKeys.UPDATE_OPS)
    g_updates = [opt for opt in ops if opt.name.startswith('generator')]
    with tf.control_dependencies(g_updates):
        g_train_opt = tf.train.AdamOptimizer(learning_rate, beta1).minimize(g_loss, var_list=g_vars)
        
    return d_train_opt, g_train_opt


"""
DON'T MODIFY ANYTHING IN THIS CELL THAT IS BELOW THIS LINE
"""
tests.test_model_opt(model_opt, tf)

Tests Passed

Neural Network Training¶

Show Output¶

Use this function to show the current output of the generator during training. It will help you determine how well the GANs is training.

"""
DON'T MODIFY ANYTHING IN THIS CELL
"""
import numpy as np

def show_generator_output(sess, n_images, input_z, out_channel_dim, image_mode):
    """
    Show example output for the generator
    :param sess: TensorFlow session
    :param n_images: Number of Images to display
    :param input_z: Input Z Tensor
    :param out_channel_dim: The number of channels in the output image
    :param image_mode: The mode to use for images ("RGB" or "L")
    """
    cmap = None if image_mode == 'RGB' else 'gray'
    z_dim = input_z.get_shape().as_list()[-1]
    example_z = np.random.uniform(-1, 1, size=[n_images, z_dim])

    samples = sess.run(
        generator(input_z, out_channel_dim, False),
        feed_dict={input_z: example_z})

    images_grid = helper.images_square_grid(samples, image_mode)
    pyplot.imshow(images_grid, cmap=cmap)
    pyplot.show()

Train¶

Implement train to build and train the GANs. Use the following functions you implemented:

model_inputs(image_width, image_height, image_channels, z_dim)
model_loss(input_real, input_z, out_channel_dim)
model_opt(d_loss, g_loss, learning_rate, beta1)

Use the show_generator_output to show generator output while you train. Running show_generator_output for every batch will drastically increase training time and increase the size of the notebook. It's recommended to print the generator output every 100 batches.

def train(epoch_count, batch_size, z_dim, learning_rate, beta1, get_batches, data_shape, data_image_mode):
    """
    Train the GAN
    :param epoch_count: Number of epochs
    :param batch_size: Batch Size
    :param z_dim: Z dimension
    :param learning_rate: Learning Rate
    :param beta1: The exponential decay rate for the 1st moment in the optimizer
    :param get_batches: Function to get batches
    :param data_shape: Shape of the data
    :param data_image_mode: The image mode to use for images ("RGB" or "L")
    """
    _, img_width, img_height, img_channels = data_shape
    real_input_img, z_input, lr = model_inputs(img_width, img_height, img_channels, z_dim)
    
    d_loss, g_loss = model_loss(real_input_img, z_input, img_channels)
    d_opt, g_opt = model_opt(d_loss, g_loss, learning_rate, beta1)
    
    steps = 0
    losses = []
    n_images = 25        
    print_every = 10
    show_every = 100
    
    with tf.Session() as sess:
        sess.run(tf.global_variables_initializer())
        for epoch_i in range(epoch_count):
            for batch_images in get_batches(batch_size):
                # Training the Model
                steps += 1 # next batch
                batch_images *= 2.0 # the image is increasing by double for each layer
                z_sample = np.random.uniform(-1, 1, (batch_size, z_dim)) # uniform distribution
                
                # performing d_opt with the feed_dict knowledge
                _ = sess.run(d_opt, feed_dict={real_input_img: batch_images, z_input: z_sample, lr: learning_rate})
                # performing g_opt with the feed_dict knowledge
                _ = sess.run(g_opt, feed_dict={z_input: z_sample, lr: learning_rate})
                
                # Check if is it time to print the Loss or just keep going (skip the print)
                if steps % print_every == 0:
                    train_loss_d = d_loss.eval({z_input: z_sample, real_input_img: batch_images})
                    train_loss_g = g_loss.eval({z_input: z_sample})
                    
                    print("Epoch {}/{}...".format(epoch_i+1, epoch_count),
                          "Discriminator Loss: {:.4f}...".format(train_loss_d),
                          "Generator Loss: {:.4f}".format(train_loss_g))
                    losses.append((train_loss_d, train_loss_g))
                
                # Showing the generator output based on show_every step
                if steps % show_every == 0:
                    show_generator_output(sess, n_images, z_input, img_channels, data_image_mode)

MNIST¶

Test your GANs architecture on MNIST. After 2 epochs, the GANs should be able to generate images that look like handwritten digits. Make sure the loss of the generator is lower than the loss of the discriminator or close to 0.

z_dim = 100 # The dimension "size" of Z
beta1 = 0.5 # The exponential decay rate for the 1st moment in the optimizer
batch_size = 64 # The size of each amount of samples per processing
learning_rate = 0.002




"""
DON'T MODIFY ANYTHING IN THIS CELL THAT IS BELOW THIS LINE
"""
epochs = 2


mnist_dataset = helper.Dataset('mnist', glob(os.path.join(data_dir, 'mnist/*.jpg')))
with tf.Graph().as_default():
    train(epochs, batch_size, z_dim, learning_rate, beta1, mnist_dataset.get_batches,
          mnist_dataset.shape, mnist_dataset.image_mode)

Epoch 1/2... Discriminator Loss: 0.0878... Generator Loss: 6.4049
Epoch 1/2... Discriminator Loss: 0.8140... Generator Loss: 1.2221
Epoch 1/2... Discriminator Loss: 0.9010... Generator Loss: 1.4534
Epoch 1/2... Discriminator Loss: 1.0379... Generator Loss: 0.6374
Epoch 1/2... Discriminator Loss: 0.9390... Generator Loss: 7.6125
Epoch 1/2... Discriminator Loss: 0.6566... Generator Loss: 0.8507
Epoch 1/2... Discriminator Loss: 0.4227... Generator Loss: 1.3187
Epoch 1/2... Discriminator Loss: 0.2860... Generator Loss: 2.0077
Epoch 1/2... Discriminator Loss: 0.7548... Generator Loss: 7.5649
Epoch 1/2... Discriminator Loss: 0.6820... Generator Loss: 2.0793

---------------------------------------------------------------------------
AttributeError                            Traceback (most recent call last)
C:\anaconda\envs\tensorflow35\lib\site-packages\IPython\core\formatters.py in __call__(self, obj)
    330                 pass
    331             else:
--> 332                 return printer(obj)
    333             # Finally look for special method names
    334             method = get_real_method(obj, self.print_method)

C:\anaconda\envs\tensorflow35\lib\site-packages\IPython\core\pylabtools.py in <lambda>(fig)
    235 
    236     if 'png' in formats:
--> 237         png_formatter.for_type(Figure, lambda fig: print_figure(fig, 'png', **kwargs))
    238     if 'retina' in formats or 'png2x' in formats:
    239         png_formatter.for_type(Figure, lambda fig: retina_figure(fig, **kwargs))

C:\anaconda\envs\tensorflow35\lib\site-packages\IPython\core\pylabtools.py in print_figure(fig, fmt, bbox_inches, **kwargs)
    119 
    120     bytes_io = BytesIO()
--> 121     fig.canvas.print_figure(bytes_io, **kw)
    122     data = bytes_io.getvalue()
    123     if fmt == 'svg':

C:\anaconda\envs\tensorflow35\lib\site-packages\matplotlib\backend_bases.py in print_figure(self, filename, dpi, facecolor, edgecolor, orientation, format, **kwargs)
   2206                     orientation=orientation,
   2207                     dryrun=True,
-> 2208                     **kwargs)
   2209                 renderer = self.figure._cachedRenderer
   2210                 bbox_inches = self.figure.get_tightbbox(renderer)

C:\anaconda\envs\tensorflow35\lib\site-packages\matplotlib\backends\backend_agg.py in print_png(self, filename_or_obj, *args, **kwargs)
    505 
    506     def print_png(self, filename_or_obj, *args, **kwargs):
--> 507         FigureCanvasAgg.draw(self)
    508         renderer = self.get_renderer()
    509         original_dpi = renderer.dpi

C:\anaconda\envs\tensorflow35\lib\site-packages\matplotlib\backends\backend_agg.py in draw(self)
    428             if toolbar:
    429                 toolbar.set_cursor(cursors.WAIT)
--> 430             self.figure.draw(self.renderer)
    431         finally:
    432             if toolbar:

C:\anaconda\envs\tensorflow35\lib\site-packages\matplotlib\artist.py in draw_wrapper(artist, renderer, *args, **kwargs)
     53                 renderer.start_filter()
     54 
---> 55             return draw(artist, renderer, *args, **kwargs)
     56         finally:
     57             if artist.get_agg_filter() is not None:

C:\anaconda\envs\tensorflow35\lib\site-packages\matplotlib\figure.py in draw(self, renderer)
   1293 
   1294             mimage._draw_list_compositing_images(
-> 1295                 renderer, self, artists, self.suppressComposite)
   1296 
   1297             renderer.close_group('figure')

C:\anaconda\envs\tensorflow35\lib\site-packages\matplotlib\image.py in _draw_list_compositing_images(renderer, parent, artists, suppress_composite)
    136     if not_composite or not has_images:
    137         for a in artists:
--> 138             a.draw(renderer)
    139     else:
    140         # Composite any adjacent images together

C:\anaconda\envs\tensorflow35\lib\site-packages\matplotlib\artist.py in draw_wrapper(artist, renderer, *args, **kwargs)
     53                 renderer.start_filter()
     54 
---> 55             return draw(artist, renderer, *args, **kwargs)
     56         finally:
     57             if artist.get_agg_filter() is not None:

C:\anaconda\envs\tensorflow35\lib\site-packages\matplotlib\axes\_base.py in draw(self, renderer, inframe)
   2397             renderer.stop_rasterizing()
   2398 
-> 2399         mimage._draw_list_compositing_images(renderer, self, artists)
   2400 
   2401         renderer.close_group('axes')

C:\anaconda\envs\tensorflow35\lib\site-packages\matplotlib\image.py in _draw_list_compositing_images(renderer, parent, artists, suppress_composite)
    136     if not_composite or not has_images:
    137         for a in artists:
--> 138             a.draw(renderer)
    139     else:
    140         # Composite any adjacent images together

C:\anaconda\envs\tensorflow35\lib\site-packages\matplotlib\artist.py in draw_wrapper(artist, renderer, *args, **kwargs)
     53                 renderer.start_filter()
     54 
---> 55             return draw(artist, renderer, *args, **kwargs)
     56         finally:
     57             if artist.get_agg_filter() is not None:

C:\anaconda\envs\tensorflow35\lib\site-packages\matplotlib\image.py in draw(self, renderer, *args, **kwargs)
    546         else:
    547             im, l, b, trans = self.make_image(
--> 548                 renderer, renderer.get_image_magnification())
    549             if im is not None:
    550                 renderer.draw_image(gc, l, b, im)

C:\anaconda\envs\tensorflow35\lib\site-packages\matplotlib\image.py in make_image(self, renderer, magnification, unsampled)
    772         return self._make_image(
    773             self._A, bbox, transformed_bbox, self.axes.bbox, magnification,
--> 774             unsampled=unsampled)
    775 
    776     def _check_unsampled_image(self, renderer):

C:\anaconda\envs\tensorflow35\lib\site-packages\matplotlib\image.py in _make_image(self, A, in_bbox, out_bbox, clip_bbox, magnification, unsampled, round_to_pixel_border)
    410 
    411                 mask = np.empty(A.shape, dtype=np.float32)
--> 412                 if A.mask.shape == A.shape:
    413                     # this is the case of a nontrivial mask
    414                     mask[:] = np.where(A.mask, np.float32(np.nan),

AttributeError: 'numpy.ndarray' object has no attribute 'mask'

<matplotlib.figure.Figure at 0x184b04c5fd0>

Epoch 1/2... Discriminator Loss: 0.4087... Generator Loss: 1.9856
Epoch 1/2... Discriminator Loss: 0.3479... Generator Loss: 2.3912
Epoch 1/2... Discriminator Loss: 1.2545... Generator Loss: 5.9889
Epoch 1/2... Discriminator Loss: 0.3329... Generator Loss: 3.9124
Epoch 1/2... Discriminator Loss: 0.2270... Generator Loss: 3.0585
Epoch 1/2... Discriminator Loss: 1.5462... Generator Loss: 7.1903
Epoch 1/2... Discriminator Loss: 0.1177... Generator Loss: 3.0669
Epoch 1/2... Discriminator Loss: 0.2827... Generator Loss: 2.3052
Epoch 1/2... Discriminator Loss: 0.8654... Generator Loss: 7.0213
Epoch 1/2... Discriminator Loss: 0.2041... Generator Loss: 3.5133

---------------------------------------------------------------------------
AttributeError                            Traceback (most recent call last)
C:\anaconda\envs\tensorflow35\lib\site-packages\IPython\core\formatters.py in __call__(self, obj)
    330                 pass
    331             else:
--> 332                 return printer(obj)
    333             # Finally look for special method names
    334             method = get_real_method(obj, self.print_method)

C:\anaconda\envs\tensorflow35\lib\site-packages\IPython\core\pylabtools.py in <lambda>(fig)
    235 
    236     if 'png' in formats:
--> 237         png_formatter.for_type(Figure, lambda fig: print_figure(fig, 'png', **kwargs))
    238     if 'retina' in formats or 'png2x' in formats:
    239         png_formatter.for_type(Figure, lambda fig: retina_figure(fig, **kwargs))

C:\anaconda\envs\tensorflow35\lib\site-packages\IPython\core\pylabtools.py in print_figure(fig, fmt, bbox_inches, **kwargs)
    119 
    120     bytes_io = BytesIO()
--> 121     fig.canvas.print_figure(bytes_io, **kw)
    122     data = bytes_io.getvalue()
    123     if fmt == 'svg':

C:\anaconda\envs\tensorflow35\lib\site-packages\matplotlib\backend_bases.py in print_figure(self, filename, dpi, facecolor, edgecolor, orientation, format, **kwargs)
   2206                     orientation=orientation,
   2207                     dryrun=True,
-> 2208                     **kwargs)
   2209                 renderer = self.figure._cachedRenderer
   2210                 bbox_inches = self.figure.get_tightbbox(renderer)

C:\anaconda\envs\tensorflow35\lib\site-packages\matplotlib\backends\backend_agg.py in print_png(self, filename_or_obj, *args, **kwargs)
    505 
    506     def print_png(self, filename_or_obj, *args, **kwargs):
--> 507         FigureCanvasAgg.draw(self)
    508         renderer = self.get_renderer()
    509         original_dpi = renderer.dpi

C:\anaconda\envs\tensorflow35\lib\site-packages\matplotlib\backends\backend_agg.py in draw(self)
    428             if toolbar:
    429                 toolbar.set_cursor(cursors.WAIT)
--> 430             self.figure.draw(self.renderer)
    431         finally:
    432             if toolbar:

C:\anaconda\envs\tensorflow35\lib\site-packages\matplotlib\artist.py in draw_wrapper(artist, renderer, *args, **kwargs)
     53                 renderer.start_filter()
     54 
---> 55             return draw(artist, renderer, *args, **kwargs)
     56         finally:
     57             if artist.get_agg_filter() is not None:

C:\anaconda\envs\tensorflow35\lib\site-packages\matplotlib\figure.py in draw(self, renderer)
   1293 
   1294             mimage._draw_list_compositing_images(
-> 1295                 renderer, self, artists, self.suppressComposite)
   1296 
   1297             renderer.close_group('figure')

C:\anaconda\envs\tensorflow35\lib\site-packages\matplotlib\image.py in _draw_list_compositing_images(renderer, parent, artists, suppress_composite)
    136     if not_composite or not has_images:
    137         for a in artists:
--> 138             a.draw(renderer)
    139     else:
    140         # Composite any adjacent images together

C:\anaconda\envs\tensorflow35\lib\site-packages\matplotlib\artist.py in draw_wrapper(artist, renderer, *args, **kwargs)
     53                 renderer.start_filter()
     54 
---> 55             return draw(artist, renderer, *args, **kwargs)
     56         finally:
     57             if artist.get_agg_filter() is not None:

C:\anaconda\envs\tensorflow35\lib\site-packages\matplotlib\axes\_base.py in draw(self, renderer, inframe)
   2397             renderer.stop_rasterizing()
   2398 
-> 2399         mimage._draw_list_compositing_images(renderer, self, artists)
   2400 
   2401         renderer.close_group('axes')

C:\anaconda\envs\tensorflow35\lib\site-packages\matplotlib\image.py in _draw_list_compositing_images(renderer, parent, artists, suppress_composite)
    136     if not_composite or not has_images:
    137         for a in artists:
--> 138             a.draw(renderer)
    139     else:
    140         # Composite any adjacent images together

C:\anaconda\envs\tensorflow35\lib\site-packages\matplotlib\artist.py in draw_wrapper(artist, renderer, *args, **kwargs)
     53                 renderer.start_filter()
     54 
---> 55             return draw(artist, renderer, *args, **kwargs)
     56         finally:
     57             if artist.get_agg_filter() is not None:

C:\anaconda\envs\tensorflow35\lib\site-packages\matplotlib\image.py in draw(self, renderer, *args, **kwargs)
    546         else:
    547             im, l, b, trans = self.make_image(
--> 548                 renderer, renderer.get_image_magnification())
    549             if im is not None:
    550                 renderer.draw_image(gc, l, b, im)

C:\anaconda\envs\tensorflow35\lib\site-packages\matplotlib\image.py in make_image(self, renderer, magnification, unsampled)
    772         return self._make_image(
    773             self._A, bbox, transformed_bbox, self.axes.bbox, magnification,
--> 774             unsampled=unsampled)
    775 
    776     def _check_unsampled_image(self, renderer):

C:\anaconda\envs\tensorflow35\lib\site-packages\matplotlib\image.py in _make_image(self, A, in_bbox, out_bbox, clip_bbox, magnification, unsampled, round_to_pixel_border)
    410 
    411                 mask = np.empty(A.shape, dtype=np.float32)
--> 412                 if A.mask.shape == A.shape:
    413                     # this is the case of a nontrivial mask
    414                     mask[:] = np.where(A.mask, np.float32(np.nan),

AttributeError: 'numpy.ndarray' object has no attribute 'mask'

<matplotlib.figure.Figure at 0x184afe84fd0>

Epoch 1/2... Discriminator Loss: 0.8154... Generator Loss: 1.4333
Epoch 1/2... Discriminator Loss: 1.4952... Generator Loss: 0.7978
Epoch 1/2... Discriminator Loss: 0.6978... Generator Loss: 3.4511
Epoch 1/2... Discriminator Loss: 0.2147... Generator Loss: 2.3866
Epoch 1/2... Discriminator Loss: 2.0010... Generator Loss: 5.3818
Epoch 1/2... Discriminator Loss: 0.9519... Generator Loss: 0.7696
Epoch 1/2... Discriminator Loss: 1.4153... Generator Loss: 3.0247
Epoch 1/2... Discriminator Loss: 1.5574... Generator Loss: 0.4268
Epoch 1/2... Discriminator Loss: 0.7372... Generator Loss: 1.4979
Epoch 1/2... Discriminator Loss: 0.3407... Generator Loss: 2.4898

---------------------------------------------------------------------------
AttributeError                            Traceback (most recent call last)
C:\anaconda\envs\tensorflow35\lib\site-packages\IPython\core\formatters.py in __call__(self, obj)
    330                 pass
    331             else:
--> 332                 return printer(obj)
    333             # Finally look for special method names
    334             method = get_real_method(obj, self.print_method)

C:\anaconda\envs\tensorflow35\lib\site-packages\IPython\core\pylabtools.py in <lambda>(fig)
    235 
    236     if 'png' in formats:
--> 237         png_formatter.for_type(Figure, lambda fig: print_figure(fig, 'png', **kwargs))
    238     if 'retina' in formats or 'png2x' in formats:
    239         png_formatter.for_type(Figure, lambda fig: retina_figure(fig, **kwargs))

C:\anaconda\envs\tensorflow35\lib\site-packages\IPython\core\pylabtools.py in print_figure(fig, fmt, bbox_inches, **kwargs)
    119 
    120     bytes_io = BytesIO()
--> 121     fig.canvas.print_figure(bytes_io, **kw)
    122     data = bytes_io.getvalue()
    123     if fmt == 'svg':

C:\anaconda\envs\tensorflow35\lib\site-packages\matplotlib\backend_bases.py in print_figure(self, filename, dpi, facecolor, edgecolor, orientation, format, **kwargs)
   2206                     orientation=orientation,
   2207                     dryrun=True,
-> 2208                     **kwargs)
   2209                 renderer = self.figure._cachedRenderer
   2210                 bbox_inches = self.figure.get_tightbbox(renderer)

C:\anaconda\envs\tensorflow35\lib\site-packages\matplotlib\backends\backend_agg.py in print_png(self, filename_or_obj, *args, **kwargs)
    505 
    506     def print_png(self, filename_or_obj, *args, **kwargs):
--> 507         FigureCanvasAgg.draw(self)
    508         renderer = self.get_renderer()
    509         original_dpi = renderer.dpi

C:\anaconda\envs\tensorflow35\lib\site-packages\matplotlib\backends\backend_agg.py in draw(self)
    428             if toolbar:
    429                 toolbar.set_cursor(cursors.WAIT)
--> 430             self.figure.draw(self.renderer)
    431         finally:
    432             if toolbar:

C:\anaconda\envs\tensorflow35\lib\site-packages\matplotlib\artist.py in draw_wrapper(artist, renderer, *args, **kwargs)
     53                 renderer.start_filter()
     54 
---> 55             return draw(artist, renderer, *args, **kwargs)
     56         finally:
     57             if artist.get_agg_filter() is not None:

C:\anaconda\envs\tensorflow35\lib\site-packages\matplotlib\figure.py in draw(self, renderer)
   1293 
   1294             mimage._draw_list_compositing_images(
-> 1295                 renderer, self, artists, self.suppressComposite)
   1296 
   1297             renderer.close_group('figure')

C:\anaconda\envs\tensorflow35\lib\site-packages\matplotlib\image.py in _draw_list_compositing_images(renderer, parent, artists, suppress_composite)
    136     if not_composite or not has_images:
    137         for a in artists:
--> 138             a.draw(renderer)
    139     else:
    140         # Composite any adjacent images together

C:\anaconda\envs\tensorflow35\lib\site-packages\matplotlib\artist.py in draw_wrapper(artist, renderer, *args, **kwargs)
     53                 renderer.start_filter()
     54 
---> 55             return draw(artist, renderer, *args, **kwargs)
     56         finally:
     57             if artist.get_agg_filter() is not None:

C:\anaconda\envs\tensorflow35\lib\site-packages\matplotlib\axes\_base.py in draw(self, renderer, inframe)
   2397             renderer.stop_rasterizing()
   2398 
-> 2399         mimage._draw_list_compositing_images(renderer, self, artists)
   2400 
   2401         renderer.close_group('axes')

C:\anaconda\envs\tensorflow35\lib\site-packages\matplotlib\image.py in _draw_list_compositing_images(renderer, parent, artists, suppress_composite)
    136     if not_composite or not has_images:
    137         for a in artists:
--> 138             a.draw(renderer)
    139     else:
    140         # Composite any adjacent images together

C:\anaconda\envs\tensorflow35\lib\site-packages\matplotlib\artist.py in draw_wrapper(artist, renderer, *args, **kwargs)
     53                 renderer.start_filter()
     54 
---> 55             return draw(artist, renderer, *args, **kwargs)
     56         finally:
     57             if artist.get_agg_filter() is not None:

C:\anaconda\envs\tensorflow35\lib\site-packages\matplotlib\image.py in draw(self, renderer, *args, **kwargs)
    546         else:
    547             im, l, b, trans = self.make_image(
--> 548                 renderer, renderer.get_image_magnification())
    549             if im is not None:
    550                 renderer.draw_image(gc, l, b, im)

C:\anaconda\envs\tensorflow35\lib\site-packages\matplotlib\image.py in make_image(self, renderer, magnification, unsampled)
    772         return self._make_image(
    773             self._A, bbox, transformed_bbox, self.axes.bbox, magnification,
--> 774             unsampled=unsampled)
    775 
    776     def _check_unsampled_image(self, renderer):

C:\anaconda\envs\tensorflow35\lib\site-packages\matplotlib\image.py in _make_image(self, A, in_bbox, out_bbox, clip_bbox, magnification, unsampled, round_to_pixel_border)
    410 
    411                 mask = np.empty(A.shape, dtype=np.float32)
--> 412                 if A.mask.shape == A.shape:
    413                     # this is the case of a nontrivial mask
    414                     mask[:] = np.where(A.mask, np.float32(np.nan),

AttributeError: 'numpy.ndarray' object has no attribute 'mask'

<matplotlib.figure.Figure at 0x184b01d1f28>

Epoch 1/2... Discriminator Loss: 0.2294... Generator Loss: 2.4727
Epoch 1/2... Discriminator Loss: 0.5400... Generator Loss: 1.8783
Epoch 1/2... Discriminator Loss: 0.5913... Generator Loss: 2.1473
Epoch 1/2... Discriminator Loss: 0.4838... Generator Loss: 1.9725
Epoch 1/2... Discriminator Loss: 0.3076... Generator Loss: 3.8957
Epoch 1/2... Discriminator Loss: 1.8004... Generator Loss: 0.5261
Epoch 1/2... Discriminator Loss: 0.7631... Generator Loss: 3.9028
Epoch 1/2... Discriminator Loss: 1.0205... Generator Loss: 1.0967
Epoch 1/2... Discriminator Loss: 0.4534... Generator Loss: 3.2543
Epoch 1/2... Discriminator Loss: 0.6494... Generator Loss: 2.5538

CelebA¶

Run your GANs on CelebA. It will take around 20 minutes on the average GPU to run one epoch. You can run the whole epoch or stop when it starts to generate realistic faces.

z_dim = z_dim
beta1 = beta1
batch_size = batch_size
learning_rate = learning_rate




"""
DON'T MODIFY ANYTHING IN THIS CELL THAT IS BELOW THIS LINE
"""
epochs = 2

celeba_dataset = helper.Dataset('celeba', glob(os.path.join(data_dir, 'img_align_celeba/*.jpg')))
with tf.Graph().as_default():
    train(epochs, batch_size, z_dim, learning_rate, beta1, celeba_dataset.get_batches,
          celeba_dataset.shape, celeba_dataset.image_mode)

Epoch 1/2... Discriminator Loss: 0.5409... Generator Loss: 6.6437
Epoch 1/2... Discriminator Loss: 1.3252... Generator Loss: 7.3785
Epoch 1/2... Discriminator Loss: 0.5452... Generator Loss: 1.6278
Epoch 1/2... Discriminator Loss: 0.2217... Generator Loss: 2.9998
Epoch 1/2... Discriminator Loss: 0.8415... Generator Loss: 4.6753
Epoch 1/2... Discriminator Loss: 0.9318... Generator Loss: 0.6426
Epoch 1/2... Discriminator Loss: 0.2589... Generator Loss: 3.0743
Epoch 1/2... Discriminator Loss: 1.7129... Generator Loss: 0.4234
Epoch 1/2... Discriminator Loss: 1.5826... Generator Loss: 4.9026
Epoch 1/2... Discriminator Loss: 1.2950... Generator Loss: 0.7303

Epoch 1/2... Discriminator Loss: 1.0192... Generator Loss: 0.5862
Epoch 1/2... Discriminator Loss: 0.6343... Generator Loss: 1.0039
Epoch 1/2... Discriminator Loss: 0.2072... Generator Loss: 2.1789
Epoch 1/2... Discriminator Loss: 3.1371... Generator Loss: 0.4011
Epoch 1/2... Discriminator Loss: 0.8817... Generator Loss: 2.8640
Epoch 1/2... Discriminator Loss: 1.0612... Generator Loss: 0.7206
Epoch 1/2... Discriminator Loss: 0.5329... Generator Loss: 1.3586
Epoch 1/2... Discriminator Loss: 0.3840... Generator Loss: 2.2172
Epoch 1/2... Discriminator Loss: 1.7202... Generator Loss: 1.3848
Epoch 1/2... Discriminator Loss: 0.6483... Generator Loss: 1.3996

Epoch 1/2... Discriminator Loss: 1.3753... Generator Loss: 1.1062
Epoch 1/2... Discriminator Loss: 0.8372... Generator Loss: 0.7834
Epoch 1/2... Discriminator Loss: 0.4149... Generator Loss: 1.4623
Epoch 1/2... Discriminator Loss: 0.3662... Generator Loss: 3.1293
Epoch 1/2... Discriminator Loss: 0.7620... Generator Loss: 2.2448
Epoch 1/2... Discriminator Loss: 3.0357... Generator Loss: 5.4549
Epoch 1/2... Discriminator Loss: 0.5821... Generator Loss: 1.2989
Epoch 1/2... Discriminator Loss: 1.1218... Generator Loss: 0.9870
Epoch 1/2... Discriminator Loss: 1.2923... Generator Loss: 1.7192
Epoch 1/2... Discriminator Loss: 0.9818... Generator Loss: 1.1606

Epoch 1/2... Discriminator Loss: 1.0629... Generator Loss: 3.1470
Epoch 1/2... Discriminator Loss: 3.8036... Generator Loss: 5.1409
Epoch 1/2... Discriminator Loss: 1.1319... Generator Loss: 0.7115
Epoch 1/2... Discriminator Loss: 0.9516... Generator Loss: 1.2629
Epoch 1/2... Discriminator Loss: 1.1288... Generator Loss: 0.6257
Epoch 1/2... Discriminator Loss: 1.0338... Generator Loss: 0.8374
Epoch 1/2... Discriminator Loss: 1.1571... Generator Loss: 0.9486
Epoch 1/2... Discriminator Loss: 1.0799... Generator Loss: 0.6960
Epoch 1/2... Discriminator Loss: 1.0433... Generator Loss: 0.6873
Epoch 1/2... Discriminator Loss: 1.2760... Generator Loss: 0.5042

Epoch 1/2... Discriminator Loss: 0.9199... Generator Loss: 1.9849
Epoch 1/2... Discriminator Loss: 1.1746... Generator Loss: 0.5873
Epoch 1/2... Discriminator Loss: 1.4513... Generator Loss: 2.4069
Epoch 1/2... Discriminator Loss: 1.2395... Generator Loss: 0.5632
Epoch 1/2... Discriminator Loss: 1.0477... Generator Loss: 0.6615
Epoch 1/2... Discriminator Loss: 0.7669... Generator Loss: 1.0555
Epoch 1/2... Discriminator Loss: 1.0309... Generator Loss: 0.6652
Epoch 1/2... Discriminator Loss: 1.5680... Generator Loss: 0.3431
Epoch 1/2... Discriminator Loss: 0.6636... Generator Loss: 1.3595
Epoch 1/2... Discriminator Loss: 2.5038... Generator Loss: 3.2951

Epoch 1/2... Discriminator Loss: 1.3835... Generator Loss: 0.5137
Epoch 1/2... Discriminator Loss: 1.5656... Generator Loss: 0.3781
Epoch 1/2... Discriminator Loss: 0.6111... Generator Loss: 1.3024
Epoch 1/2... Discriminator Loss: 0.6167... Generator Loss: 1.2104
Epoch 1/2... Discriminator Loss: 1.4702... Generator Loss: 0.5422
Epoch 1/2... Discriminator Loss: 0.9288... Generator Loss: 1.1823
Epoch 1/2... Discriminator Loss: 0.8821... Generator Loss: 1.7834
Epoch 1/2... Discriminator Loss: 1.0103... Generator Loss: 2.5550
Epoch 1/2... Discriminator Loss: 1.2324... Generator Loss: 2.7544
Epoch 1/2... Discriminator Loss: 1.4156... Generator Loss: 0.4468

Epoch 1/2... Discriminator Loss: 1.1114... Generator Loss: 0.6631
Epoch 1/2... Discriminator Loss: 0.9133... Generator Loss: 0.9225
Epoch 1/2... Discriminator Loss: 0.7469... Generator Loss: 2.0840
Epoch 1/2... Discriminator Loss: 1.9063... Generator Loss: 3.5590
Epoch 1/2... Discriminator Loss: 0.9908... Generator Loss: 0.9781
Epoch 1/2... Discriminator Loss: 0.8639... Generator Loss: 1.1782
Epoch 1/2... Discriminator Loss: 0.8045... Generator Loss: 1.2929
Epoch 1/2... Discriminator Loss: 0.8037... Generator Loss: 1.2021
Epoch 1/2... Discriminator Loss: 1.8171... Generator Loss: 0.2723
Epoch 1/2... Discriminator Loss: 1.5314... Generator Loss: 0.4016

Epoch 1/2... Discriminator Loss: 1.6685... Generator Loss: 2.6423
Epoch 1/2... Discriminator Loss: 1.5163... Generator Loss: 0.4920
Epoch 1/2... Discriminator Loss: 1.5247... Generator Loss: 3.4482
Epoch 1/2... Discriminator Loss: 0.6517... Generator Loss: 1.2485
Epoch 1/2... Discriminator Loss: 0.7217... Generator Loss: 1.3765
Epoch 1/2... Discriminator Loss: 0.8033... Generator Loss: 0.9621
Epoch 1/2... Discriminator Loss: 2.2713... Generator Loss: 0.1535
Epoch 1/2... Discriminator Loss: 0.6873... Generator Loss: 1.5382
Epoch 1/2... Discriminator Loss: 1.0817... Generator Loss: 0.7517
Epoch 1/2... Discriminator Loss: 1.1375... Generator Loss: 0.6480

Epoch 1/2... Discriminator Loss: 0.8254... Generator Loss: 1.2566
Epoch 1/2... Discriminator Loss: 0.9945... Generator Loss: 0.9732
Epoch 1/2... Discriminator Loss: 1.2296... Generator Loss: 0.5966
Epoch 1/2... Discriminator Loss: 1.2268... Generator Loss: 0.7802
Epoch 1/2... Discriminator Loss: 1.1435... Generator Loss: 0.5699
Epoch 1/2... Discriminator Loss: 1.0861... Generator Loss: 0.6385
Epoch 1/2... Discriminator Loss: 0.7214... Generator Loss: 1.4172
Epoch 1/2... Discriminator Loss: 0.8062... Generator Loss: 1.4388
Epoch 1/2... Discriminator Loss: 1.3215... Generator Loss: 0.5128
Epoch 1/2... Discriminator Loss: 0.7705... Generator Loss: 1.1615

Epoch 1/2... Discriminator Loss: 1.1346... Generator Loss: 0.8681
Epoch 1/2... Discriminator Loss: 1.0953... Generator Loss: 0.7579
Epoch 1/2... Discriminator Loss: 0.9095... Generator Loss: 0.9349
Epoch 1/2... Discriminator Loss: 0.9948... Generator Loss: 1.5497
Epoch 1/2... Discriminator Loss: 1.0282... Generator Loss: 1.9258
Epoch 1/2... Discriminator Loss: 1.3438... Generator Loss: 0.4900
Epoch 1/2... Discriminator Loss: 1.3134... Generator Loss: 0.4605
Epoch 1/2... Discriminator Loss: 0.8145... Generator Loss: 1.1744
Epoch 1/2... Discriminator Loss: 1.0510... Generator Loss: 0.6364
Epoch 1/2... Discriminator Loss: 1.1008... Generator Loss: 0.6483

Submitting This Project¶

When submitting this project, make sure to run all the cells before saving the notebook. Save the notebook file as "dlnd_face_generation.ipynb" and save it as a HTML file under "File" -> "Download as". Include the "helper.py" and "problem_unittests.py" files in your submission.